1.Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China 2.University of Science and Technology of China, Hefei 230026, China 3.Key Laboratory of Optical Calibration and Characterization, Chinese Academy of Sciences, Hefei 230031, China
Fund Project:Project supported by the National Key Research and Development Program of China (Grant No. 2016YFE0201400), the Common Key Technology Project for Satellite Application of China (Grant No. 30-Y20A010-9007-17/18), the National High Resolution Major Special Project of China (Grant No. GFZX04011805), and the Key Project of Hefei Research Institute of Chinese Academy of Sciences (Grant No. Y73H9P1801).
Received Date:18 August 2018
Accepted Date:20 October 2018
Available Online:01 January 2019
Published Online:20 January 2019
Abstract:The division-of-amplitude full Stokes simultaneous polarization imaging system has prominent merits, such as real time, high spatial resolution, high precision, etc. The development of the division-of-amplitude full Stokes simultaneous polarization imaging system has a high application value. The division-of-amplitude full Stokes simultaneous polarization imaging system uses polarization beam splitters, a half wave plate (HWP) and a quarter wave plate (QWP) to modulate the incident Stokes vector into four intensity images. Using the four intensity images, the incident Stokes vector can be analyzed. In the system, the phase delay errors of the HWP and the QWP have a direct influence on the measurement accuracy of the incident Stokes vector. A Stokes vector measurement error equation containing the phase delay errors of the HWP and the QWP is established. When there are the phase delay errors of the HWP and the QWP in the system, the Stokes vector measurement errors of the unpolarized light, 0° liner polarized light, 90° liner polarized light, 45° liner polarized light, 135° liner polarized light, right circularly polarized light and left circularly polarized light are analyzed. A method of solving the Stokes vector measurement error of incident light with any polarization state is given. When the Stokes vectors with different degrees of polarization (DOPs) are used as the incident light, the simulation results show that both the Stokes vector measurement error and the DOP measurement error increase with the DOP of incident light increasing. Therefore, we select the polarization measurement accuracy to evaluate the system when the DOP of incident light equals 1. To ensure that the polarization measurement accuracy of the system is within 2%, the phase delay error of the HWP should be within ±1.6° and the phase delay error of the QWP should be within ±0.5°. The analysis results of the phase delay errors of the HWP and the QWP are of great significance for improving the polarization measurement accuracy of the division-of-amplitude full Stokes simultaneous polarization imaging system, and also provide important theoretical guidance in designing and developing the system. Keywords:polarization imaging/ wave plate/ phase delay error/ Stokes vector
全文HTML
--> --> --> -->
2.1.偏振成像系统工作原理
分振幅型全Stokes同时偏振成像系统原理如图1所示, 入射光通过光学镜头后, 被部分偏振分束器(partial polarization beam splitter, PPBS)分成两路, 一路是从PPBS透射的光束, 被HWP调制后由PBS1分离为P偏振光和S偏振光, 从PBS1透射的P偏振光由CCD1接收, 反射的S偏振光由CCD2接收; 另一路是从PPBS反射的光束, 被QWP调制后由PBS2分离为P偏振光和S偏振光, 从PBS2透射的P偏振光由CCD3接收, 反射的S偏振光由CCD4接收. 图 1 分振幅型全Stokes同时偏振成像系统原理图 Figure1. Scheme of the division-of-amplitude full Stokes simultaneous polarization imaging system.
式中${{{S}}_n}$为邦加球球面上或球内的第n个入射光采样点的归一化Stokes矢量, P为偏振度, ${\chi _n}$和${\psi _n}$分别为第n个入射光采样点${{{S}}_n}$的椭圆度角和偏振角, ${N_\chi }$表示从邦加球的南极到北极围绕球体采样的圆周个数, ${N_\psi }$表示每个采样圆周上的采样点数. 令(12)式中的P = 1, ${N_\chi }$ = 20, ${N_\psi }$ = 50, 则图3为邦加球球面上1000个不同偏振态的Stokes矢量的三维分布和Stokes参数的数值分布. 图 3 邦加球球面上选取1000个不同偏振态的Stokes矢量的 (a) 三维分布和(b) Stokes参数的数值分布 Figure3. (a) 3D distribution and (b) stokes parameters values of 1000 Stokes vectors different degrees of polarization selected on the Poincaré sphere.
以表1中的系统设计参数作为仿真参数, 以图3中的1000个Stokes矢量作为入射光采样点, 则图4(a)和图4(b)分别为仅存在1°的HWP相位延迟误差时和仅存在1°的QWP相位延迟误差时1000个入射光采样点的Stokes参数${{{S}}_0}$?${{{S}}_3}$测量误差. 图4中横坐标为入射光采样点从邦加球南极到北极的采样序数, 纵坐标为Stokes参数测量误差. 图 4 1000个邦加球球面上的入射光采样点的Stokes参数测量误差 (a) 仅存在1°的 HWP相位延迟误差; (b) 仅存在1°的QWP相位延迟误差 Figure4. The measurement errors of Stokes parameters of 1000 incident light sampling points selected on the Poincaré sphere is simulated: (a) There is only 1° phase delay error of HWP; (b) there is only 1° phase delay error of QWP in the system.
以表1中的系统设计参数作为仿真参数, 以图3中的1000个邦加球球面上(偏振度P均为1)的Stokes矢量作为入射光采样点, 则图6(a)和图6(b)分别为仅存在1° 的HWP相位延迟误差时和仅存在1° 的QWP相位延迟误差时1000个邦加球球面上的入射光采样点的偏振度测量误差. 图 6 1000个邦加球球面上的入射光采样点的偏振度测量误差 (a) 仅存在1°的HWP相位延迟误差; (b) 仅存在1°的QWP相位延迟误差 Figure6. The measurement errors of DOP of 1000 incident light sampling points selected on the Poincaré sphere is simulated: (a) There is only 1° phase delay error of HWP; (b) there is only 1° phase delay error of QWP in the system.
表2系统偏振度测量精度${\rm{acc}}\_P(\sigma ,\delta ,P {\rm{ = }} 1)$随HWP相位延迟误差$\sigma $和QWP相位延迟误差$\delta $的变化关系 Table2.Variation relation of measure accuracy ${\rm{acc}}\_P(\sigma ,\delta ,P {\rm{ = }} 1)$ of DOP with the phase delay error $\sigma $ of HWP and the phase delay error $\delta $ of QWP.
$\sigma$
$\delta $
?1.0°
?0.9°
?0.6°
?0.5°
0°
0.5°
0.6°
0.9°
1.0°
?3.2°
4.01%
4.01%
4.01%
4.01%
4.01%
4.01%
4.01%
4.01%
4.01%
?3.1°
3.88%
3.88%
3.88%
3.88%
3.88%
3.88%
3.88%
3.88%
3.88%
?1.7°
3.48%
3.14%
2.11%
2.11%
2.11%
2.11%
2.11%
3.14%
3.48%
?1.6°
3.48%
3.14%
2.09%
1.99%
1.99%
1.99%
2.09%
3.14%
3.48%
?0.5°
3.48%
3.14%
2.09%
1.74%
0.62%
1.74%
2.09%
3.14%
3.48%
0°
3.48%
3.14%
2.09%
1.74%
0
1.74%
2.09%
3.14%
3.48%
0.5°
3.48%
3.14%
2.09%
1.74%
0.62%
1.74%
2.09%
3.14%
3.48%
1.6°
3.48%
3.14%
2.09%
1.99%
1.99%
1.99%
2.09%
3.14%
3.48%
1.7°
3.48%
3.14%
2.11%
2.11%
2.11%
2.11%
2.11%
3.14%
3.48%
3.1°
3.88%
3.88%
3.88%
3.88%
3.88%
3.88%
3.88%
3.88%
3.88%
3.2°
4.01%
4.01%
4.01%
4.01%
4.01%
4.01%
4.01%
4.01%
4.01%
表3系统偏振测量精度${\rm{acc}}\_{{{S}}^{\left( {\sigma ,\delta } \right)}}$随HWP相位延迟误差$\sigma $和QWP相位延迟误差$\delta $的变化关系 Table3.Variation relation of system polarization measurement accuracy ${\rm{acc}}\_{{{S}}^{\left( {\sigma ,\delta } \right)}}$ with the phase delay error $\sigma $ of HWP and the phase delay error $\delta $ of QWP.